Played a Numa organ tonight

[HammondDave]

Just got back from the gig. I had to reset the Numa twice during the last two songs of a 90 minute set. A note hung in both instances when doing multiple glissandos.

 

See what happens when you diss an instrument? That Numa read your complaints about it's latency and said: "So you complain about my tardiness??? Lets see how you like THIS!!!! CRASHHH!!!!"...

 

 

[CrimsonianKing]

LOL

[LX88]

Just got done playing my Numa out in the back yard on a warm summer night....

 

The real latency issue is not in this board. It is between my brain and my fingers.

 

[Jim Alfredson]

Is there a PDF manual available anywhere? I'd like to get more familiar with the Numa before tonight's gig.

[nickd]

I think this is what you need:

http://www.synthax.de/media/files/studiologic/NumaOrgan-Manual-1_3-Web.pdf

[Jim Alfredson]

Thanks, dood!

[CEB]

The gas pedal is so important to me it is the reason why other vendors can't touch Hammond Suzuki. I was really close to buying a Nord but Clavia or no one else can currently touch the Hammond EXP-100F.

[JeffLearman]

The real latency issue is not in this board. It is between my brain and my fingers.

You and me both. I tolerate a lot more latency on hammond than on piano, maybe because I'm a piano player first. Even then I tolerate more than most, probably for the same reason as you. :-|

 

Jim, I'm surprised you don't carry around your own expression pedal. Doesn't anyone make one with a compatibility switch? I believe there are only two types (hi-Z and higher-Z).

 

BTW, maybe that's why the pedal isn't working well: it's the wrong type. My guess was that they should be interchangeable (since it's just a voltage divider, who cares what the resistance values are)? But folks who've tried mix & match say it causes problems, and experience trumps theory, especially in this case since I'm no electronics whiz.

[HammondDave]

The real latency issue is not in this board. It is between my brain and my fingers.

You and me both. I tolerate a lot more latency on hammond than on piano, maybe because I'm a piano player first. Even then I tolerate more than most, probably for the same reason as you. :-|

 

Jim, I'm surprised you don't carry around your own expression pedal. Doesn't anyone make one with a compatibility switch? I believe there are only two types (hi-Z and higher-Z).

 

BTW, maybe that's why the pedal isn't working well: it's the wrong type. My guess was that they should be interchangeable (since it's just a voltage divider, who cares what the resistance values are)? But folks who've tried mix & match say it causes problems, and experience trumps theory, especially in this case since I'm no electronics whiz.

 

 

I know that the XK3c has quite a few menu parameters for expression pedals. When my H/Z pedal failed I used one of my Korg pedals... Once I tweaked the parameters it worked very well... Does the Numa have the same flexibility?

[Joe Muscara]

The gas pedal is so important to me it is the reason why other vendors can't touch Hammond Suzuki. I was really close to buying a Nord but Clavia or no one else can currently touch the Hammond EXP-100F.

I wonder what the EXP-50 (for the SK-1/2) is like and if they're available yet.

[Jim Alfredson]

According to the manual, the Numa wants a linear pedal. My guess is the pedal they have supplied is exponential. You can use both with the XK3/c by changing it in the menu system. No menu on the Numa that I can find (no screen).

 

They supplied another pedal today but it's the same brand and model so it does the same thing. Oh well. Last gig with the Numa tonight. Tomorrow I get the KeyB again.

[HammondDave]

According to the manual, the Numa wants a linear pedal. My guess is the pedal they have supplied is exponential. You can use both with the XK3/c by changing it in the menu system. No menu on the Numa that I can find (no screen).

 

They supplied another pedal today but it's the same brand and model so it does the same thing. Oh well. Last gig with the Numa tonight. Tomorrow I get the KeyB again.

 

Man, that would just piss me off!

 

For some reason StudioLogic believes that the musician should have no choice in tweaking the sound or the functions of the Numa. Maybe you need a certified letter from "The Pope" to make any changes.

[LX88]

I don't know if that is Studiologic that gives you no choice in tweakability. I kind of think that was Elvio's decision, based on the conversations I had with him. Just like the Key B is not really very tweakable.

 

I have tried The Roland EV-5 pedal , the Yamaha FC-7.... they all worked for me.

 

As far as this whole latency thing goes I am not trying to be critical or cause a nasty debate but when I played my Numa last night I didn't experience anything that felt delayed. And I do love the sound.

 

No it is not perfect, but at home I can work with it. I am not so sure what it would be like on the road through whatever system was supplied.

 

So If I were on the road I would go for something more tweakable certainly. And the Hammond SK-1 is looking like it might be a good overall choice. We are waiting to see how that pans out.

 

But I am not sorry I got the Numa. It is working out well with my own amp system and I love it for practicing.

 

The chorus vibrato and the good leslie sim are just so nice to have right inside the board. The Numa sim does not sound colored to me , unlike many of the other sims that had been out previously. And the chorus just never makes me feel like I am lacking something.

 

So to me ,still.... the plusses outweigh the minuses.

[Jim Alfredson]

Just got done with the last gig on this tour with the Numa. It's just not for me. I cannot get past the disconnect between the keys and the sound. There's too much delay. I really notice it when I'm doing staccato rhythmic chords. It's almost like it delays ever so slightly even more the faster I play. It's weird.

 

Also, I miss the upper end of the XK3. The Numa is very mellow. It would be great for jazz. I like the overdrive, too. But I could never get it to jump out at me through the monitors.

 

It's great that the FC-7 works. Next time I come to Europe, I'll pack mine with me, just in case.

 

LX88, I'm glad you're happy with yours. I won't cringe if I run into another one as backline again, but it certainly won't be my first choice.

 

I'm curious if another Numa owner would do a latency check and post the results.

 

 

This all just makes me pine for the SK1 even more. To be able to have something I know with me at all times... well, that would be heaven. Beyond the organ, tonight I had a Roland RD700sx. I think the "sx" stands for "sux". Man, that thing was awful. Awful action, mediocre piano, horrendous electric pianos. I should've hooked up my laptop but I didn't want to freak out the sound guys. Next time I run into one of those, I'm just going to run Pianoteq on the laptop, though. Sound guys be damned.

[Al Coda]

I cannot get past the disconnect between the keys and the sound. There's too much delay. I really notice it when I'm doing staccato rhythmic chords. It's almost like it delays ever so slightly even more the faster I play. It's weird.

 

thx for your reports !

Damn, that´s exactly what drives me nuts w/ latency too.

Now I´m happy I didn´t pull the trigger until now.

As a general rule,- I can live w/ compromises in sound much better than dealing w/ that amount of latency which acts like a brake while pushing the pedal to the metal.

 

A.C.

 

 

[HammondDave]

So Jim... Is the latency on the Numa worse than using a controller and the VB3?

[JeffLearman]

With the VB3 you can at least configure the computer for low audio latency.

 

Al Coda posted a link to a test for MIDI latency (ignoring the audio part), and the data at that site showed higher numbers than I would have guessed, but with a lot of variation depending on type of MIDI adaptor. No doubt the computer and its configuration matters a lot as well.

[MusicaL]

I've experienced nothing of the sort of latency Jim measured. This is non scientific but there's nothing disconnected with my Numa. It plays as well as any other instrument I own, including my E3 and Stage compact. Could it be an

older unit with a non updated OS?

 

aL

[RGH]

I would agree with Al. I have had my Numa since late March, and it seems to react just as well as any of my other keyboards.

I am sure that some of the units in Europe have been out there longer, though.

As for the expression pedal issues, I sometimes use my Roland EV5, with fairly good success, but the best pedal has been my Roland EV7,that I formerly used with my VK7.

[Jim Alfredson]

So Jim... Is the latency on the Numa worse than using a controller and the VB3?

 

Yes. Well, of course it depends on your computer, but with my laptop I can get the latency down to around 6ms with VB3.

 

I've experienced nothing of the sort of latency Jim measured. This is non scientific but there's nothing disconnected with my Numa. It plays as well as any other instrument I own, including my E3 and Stage compact. Could it be an

older unit with a non updated OS?

 

aL

 

It could be. That would be great news if so.

 

Rather than go by subjective opinion, why not test the Numa using the same setup I described? Put a mic above a key, connect that mic to your DAW. Take a mono output from the Numa, connect that to your DAW. Hit a key and record both the mic and the direct, then zoom in (with your DAW's timescale set to samples) and measure the delay.

[Al Coda]

Yes. Well, of course it depends on your computer, but with my laptop I can get the latency down to around 6ms with VB3.

 

So, you measured the VB3 runnung on the laptop the same way you´re describing below ?

 

B.t.w.,- with both of my rackmount DAWs,- both outdated, Asus mobos, Pentium 4 processors, one machine a single core 2.8Ghz, the other a dual core 3.82 GHz,- 5.8ms is a safe setting for me to run not only VB3 but several plugins simultaneously without clicks, pops or crackling.

I can go down to 3ms running 1 plugin only w/ the dual core.

 

But,- the dual core operates w/ a USB MIDI interface and the single core machine w/ a LTP port interface,- the single core machine feels tighter and more responsive and my guess is,- it´s because non USB MIDI !

OTOH, I can run more plugins on the dual core machine.

 

 

Rather than go by subjective opinion, why not test the Numa using the same setup I described? Put a mic above a key, connect that mic to your DAW. Take a mono output from the Numa, connect that to your DAW. Hit a key and record both the mic and the direct, then zoom in (with your DAW's timescale set to samples) and measure the delay.

 

Just to be sure,- you record the noise of the key when it´s hit by the finger w/ the mic ?

You record the audio signal from a line out of the instrument going into the line in of your interface on your computer´s harddrive ?

 

If yes, how much latency you record w/ the mic signal alone ?

There´s audio input latency w/ any computer (ASIO Input latency and AD conversion) which you might have to subtract from the result.

Did you consider this ?

 

As I mentioned in another thread, it´s very hard for a user to measure the latency of hardware keyboards.

It´s some kind of a puzzle for accuracy.

 

1.)

After a keystroke and key contact closed,- how long needs the processor to process MIDI note ON and velocity value triggering the tone generator/engine to produce a sound which then will appear digitally at the DA input and appears analog at the audio output as a electric signal (waveform).

That´s the real latency of that keyboard instrument for only ONE note.

 

2.)

You need some test equipment do do such a test and have to be able to read the schematics (if available) of that instrument to know where to measure what inside the keyboards circuitry.

 

Well, that´s a unacceptable task for the average musician IMO.

So, you better trust your feel and ears.

If you feel disconnected,- don´t buy it.

 

But, if someone has the test equipment and the skills and is willing to do such tests, it would be very informative !

 

A.C.

[JeffLearman]

Just to be sure,- you record the noise of the key when it´s hit by the finger w/ the mic ?

You record the audio signal from a line out of the instrument going into the line in of your interface on your computer´s harddrive ?

 

If yes, how much latency you record w/ the mic signal alone ?

There´s audio input latency w/ any computer (ASIO Input latency and AD conversion) which you might have to subtract from the result. Did you consider this ?

Since both signals (keyboard's audio out and mic by key) are recorded using the same interface, the two tracks will be synchronized regardless of any audio latency.

 

As I've mentioned in other threads, this is the simple and easy way to measure total round-trip latency regardless of the type of keyboard instrument. For the purposes here, it's better than the standard lab bench method. There is one uncontrolled variable, though: the time between hearing the knuckle strike the key and the key contact closing. If anyone has a good idea on measuring that, I'd love to hear it. In any case, it's minimized by striking with a good swift whack. I expect it's less than a millisecond.

 

The above method is accurate except for the one unmeasured variable, and includes every source of latency except for transmission in the air -- i.e., it's equivalent to monitoring with headphones. We don't need to measure this; we can estimate it, and it's beside the main point here anyway.

 

No special equipment or skills required, other than using a DAW.

 

But like Al Coda, I would be interested to know what the latency is for Jim's VB3 measured the same way. Usually what people report for VSTi latency is the number that appears in the buffer configuration popup, which accounts for only the audio buffering latency. This is usually the largest factor, but there are other significant ones.

 

I'd venture it's less than 12 ms, but how much less?

[Joe Muscara]

If it's possible, I would also measure the MIDI data (Note On) as well, especially when testing software instruments such as VB3.

[JMcS]

This product claims to be latency free but it is a HOAX:

 

HOAX B3

[MusicaL]

Jim,

 

I'll give that latency test a go. Not sure I am a sufficient techie to pull it off, but, I'll try it and see what I come up with.

 

Does the mic capture the noise of your finger hitting the key or the noise of the key at the bottom of its travel (or both)? Is there a significant interval between the the time your finger hits the key and the key contact closes that should be accounted for?

 

aL

[analogholic]

Yes. Well, of course it depends on your computer, but with my laptop I can get the latency down to around 6ms with VB3.

 

 

Does 6ms feel good or does it only feel better than the Numa?

 

[Al Coda]

Since both signals (keyboard's audio out and mic by key) are recorded using the same interface, the two tracks will be synchronized regardless of any audio latency.

 

Yes, that´s right, Jeff,- but I wanted to be sure, Jim subtracted the audio input latency on BOTH tracks for reporting the measured latency in milliseconds.

 

If he did, the result is amazingly bad.

If he didn´t, that means the real latency would be less but dis-connectivity already starts at lower latency values for a player not only playing chords and pads but explosive runs and complex rhythmic patterns !

 

A.C.

[Al Coda]

 

Does 6ms feel good or does it only feel better than the Numa?

 

I myself, I tolerate 6ms for savety/headroom reasons on my DAW if I play and record MIDI.

There are also some VSTi which seem to introduce some kind of additional latency compared to other VSTi.

This becomes obvious if you play 6 voice block chords repeatedly p.ex..

 

I´d like to come down below 3ms w/ several VSTi and FX loaded into a VST rack.

 

Example and no need to do measurements:

 

Audio/MIDI PCI card = Scope 15DSP PCI

 

6 synths (monophonic and polyphonic) run on Scope DSPs incl. a mixer,- the synths triggered by Scope PCI MIDI.

Latency is a few samples,- no audible difference to hardware.

 

In addition,- some VSTi loaded in a VST host (Reaper or Cantabile lite), the host using Scope ASIO driver (buffer size 6-7 ms @44.1K), ASIO Input module connected to Scope mixer.

VSTis triggered by a separate 8x8 MIDI interface connected to the computer mobo (can be USB or LTP,- I tested both).

Latency pretty much feelable and audible against the Scope synths running on DSP and more audible if compared to the analog output signal of the MIDI controller keyboard (Rhodes MK80).

 

The reported latency of the NUMA makes me wonder because NUMA runs on Sharc DSP IIRR,- like Scope.

IF I remember right and the latency of NUMA is reported correctly, it can´t be the processor,- it must be the code of the NUMA and/or a MIDI issue.

 

A.C.

[Al Coda]

This product claims to be latency free but it is a HOAX:

 

HOAX B3

 

Interesting and amazing !

 

A.C.

[KRK]

ciao Jim, ciao to all!

:-)))))

 

(it's Marco from Italy)

 

 

never had such latency issue with my keybduo.

are you using the new dsp version?

 

I can remember (but do not ask me how to do it :-)))) ) that there is a way, with the keybduo, to make the keyboard "respond" in two different ways.

one is the standard midi touch;

the other way ("organ mode") allows the keyboard to make the generator sound as soon as you touch the key, as on a vintage hammond.

 

you can shift from a mode to the other using a combination of keys (maybe your keyb was setted in a "strange" way?)

 

 

 

concerning the Numa: yes, I can confirm that the "feeling" of the keyboard is different from the keybduo.

I do not experienced latency issues (but, for the Numa, unfortunately I just used it for a couple of hours during my home-made test - see it on youtube - user marcoballa, as you remeber :-))))) , just to show that the sound engine is the same of the keybduo)

 

 

take care

 

Marco

 

 

OT_01:

I'm so sorry I was on the seaside while you were playing around here in Italy!!!!!!!

 

 

OT_02:

next week a friend of mine will receive his SK2, so I'll ask him to give it to me for a weekend :-))))

I'll make a new video with high quality audio, with his simulator, with the Ventilator and with the 3300 leslie)